pip install --upgrade plotly

Requirement already satisfied: plotly in /usr/local/lib/python3.12/dist-packages (5.24.1)
Collecting plotly
  Downloading plotly-6.3.0-py3-none-any.whl.metadata (8.5 kB)
Requirement already satisfied: narwhals>=1.15.1 in /usr/local/lib/python3.12/dist-packages (from plotly) (2.3.0)
Requirement already satisfied: packaging in /usr/local/lib/python3.12/dist-packages (from plotly) (25.0)
Downloading plotly-6.3.0-py3-none-any.whl (9.8 MB)
   ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 9.8/9.8 MB 52.4 MB/s eta 0:00:00
Installing collected packages: plotly
  Attempting uninstall: plotly
    Found existing installation: plotly 5.24.1
    Uninstalling plotly-5.24.1:
      Successfully uninstalled plotly-5.24.1
Successfully installed plotly-6.3.0

!pip install squarify

Collecting squarify
  Downloading squarify-0.4.4-py3-none-any.whl.metadata (600 bytes)
Downloading squarify-0.4.4-py3-none-any.whl (4.1 kB)
Installing collected packages: squarify
Successfully installed squarify-0.4.4

import numpy as np
import pandas as pd
import os
import squarify
import matplotlib.pyplot as plt
import plotly.express as px
import plotly.graph_objects as go
import seaborn as sns
from sklearn.ensemble import RandomForestClassifier
from sklearn.model_selection import train_test_split
from sklearn.metrics import accuracy_score
from sklearn.preprocessing import LabelEncoder

hr_data = pd.read_csv("/content/HR_Data_MNC.csv")

hr_data.head(10)

hr_data.shape

(86754, 12)

print(hr_data.info(), "\n")
print(hr_data.isnull().sum())

<class 'pandas.core.frame.DataFrame'>
RangeIndex: 86754 entries, 0 to 86753
Data columns (total 12 columns):
 #   Column              Non-Null Count  Dtype  
---  ------              --------------  -----  
 0   Unnamed: 0          86754 non-null  int64  
 1   Employee_ID         86754 non-null  object 
 2   Full_Name           86754 non-null  object 
 3   Department          86754 non-null  object 
 4   Job_Title           86754 non-null  object 
 5   Hire_Date           86753 non-null  object 
 6   Location            86753 non-null  object 
 7   Performance_Rating  86753 non-null  float64
 8   Experience_Years    86753 non-null  float64
 9   Status              86753 non-null  object 
 10  Work_Mode           86753 non-null  object 
 11  Salary_INR          86753 non-null  float64
dtypes: float64(3), int64(1), object(8)
memory usage: 7.9+ MB
None 

Unnamed: 0            0
Employee_ID           0
Full_Name             0
Department            0
Job_Title             0
Hire_Date             1
Location              1
Performance_Rating    1
Experience_Years      1
Status                1
Work_Mode             1
Salary_INR            1
dtype: int64

hr_data.count()

hr_data.columns

Index(['Unnamed: 0', 'Employee_ID', 'Full_Name', 'Department', 'Job_Title',
       'Hire_Date', 'Location', 'Performance_Rating', 'Experience_Years',
       'Status', 'Work_Mode', 'Salary_INR'],
      dtype='object')

hr_data.columns = hr_data.columns.str.upper()

hr_data

# Rename the column
hr_data.rename(columns={'UNNAMED: 0': 'SI.'}, inplace=True)

# Verify the change
hr_data.head()

hr_data.count()

hr_data.describe()

# Check unique values in STATUS column
hr_data['STATUS'].value_counts()

import matplotlib.pyplot as plt
from matplotlib.ticker import ScalarFormatter

status_counts = hr_data['STATUS'].value_counts()
colors = ['skyblue', 'orange', 'blue', 'red']

plt.figure(figsize=(8,5))
bars = plt.bar(status_counts.index, status_counts.values, color=colors, edgecolor='black')

# Add values on top
for bar in bars:
    height = bar.get_height()
    plt.text(bar.get_x() + bar.get_width()/2, height + 2500, f'{height:,}', ha='center', va='bottom')

# Format y-axis to show normal numbers instead of scientific notation
plt.gca().yaxis.set_major_formatter(ScalarFormatter())
plt.title('Distribution of Employee Status')
plt.xlabel('Status')
plt.ylabel('Number of Employees in (mln)')
plt.xticks(rotation=0)
plt.show()

import plotly.express as px

# Prepare the value counts
status_counts = hr_data['STATUS'].value_counts().reset_index()
status_counts.columns = ['Status', 'Count']  # Rename columns properly

# Plot
fig = px.bar(status_counts,
             x='Status',
             y='Count',
             title='Distribution of Employee Status',
             color='Status',
             text='Count')  # Show counts on bars

fig.show()

# Check unique work modes and counts
hr_data['WORK_MODE'].value_counts()

import matplotlib.pyplot as plt
import pandas as pd

# Example dataset (you can replace with your actual data)
data = ['Remote', 'On-site', 'Hybrid', 'Remote', 'Remote', 'Hybrid', 'On-site']

# Count occurrences of each work mode
work_mode_counts = pd.Series(data).value_counts()

# Define colors (adjust if categories differ)
colors = ['skyblue', 'orange', 'pink']

# Plot pie chart
plt.figure(figsize=(7,7))
plt.pie(work_mode_counts,
        labels=work_mode_counts.index,
        autopct='%1.1f%%',   # Show percentage with 1 decimal
        startangle=140,
        colors=colors,
        shadow=True)

plt.title('Distribution of Work Modes')
plt.show()

import matplotlib.pyplot as plt
from matplotlib.ticker import ScalarFormatter

# Count of each work mode
work_mode_counts = hr_data['WORK_MODE'].value_counts()

# Define colors for each work mode
colors = ['skyblue', 'orange', 'green']  # Adjust based on number of categories

# Plot
plt.figure(figsize=(8,5))
bars = plt.bar(work_mode_counts.index, work_mode_counts.values, color=colors, edgecolor='black')

# Add values on top of each bar
for bar in bars:
    height = bar.get_height()
    plt.text(bar.get_x() + bar.get_width()/2, height + 5000, f'{height:,}', ha='center', va='bottom')

# Format y-axis to show normal numbers instead of scientific notation
plt.gca().yaxis.set_major_formatter(ScalarFormatter())

plt.title('Distribution of Work Modes')
plt.xlabel('Work Mode')
plt.ylabel('Number of Employees')
plt.xticks(rotation=0)
plt.show()

# Count of employees in each department
department_counts = hr_data['DEPARTMENT'].value_counts()

# Display counts
print(department_counts)

DEPARTMENT
IT            25934
Sales         17213
Operations    13181
Marketing     10429
Finance        8732
HR             7020
R&D            4245
Name: count, dtype: int64

plt.figure(figsize=(12,6))
bars = plt.barh(department_counts.index, department_counts.values, color=plt.cm.tab20.colors, edgecolor='black')

# Add values
for bar in bars:
    width = bar.get_width()
    plt.text(width + 5000, bar.get_y() + bar.get_height()/2, f'{width:,}', va='center')

plt.gca().xaxis.set_major_formatter(ScalarFormatter())
plt.xlabel('Number of Employees')
plt.ylabel('Department')
plt.title('Number of Employees per Department')
plt.show()

import matplotlib.pyplot as plt
import squarify

# Sort departments by size (optional, makes treemap look better)
department_counts_sorted = department_counts.sort_values(ascending=False)

# Create color palette (enough colors for all departments)
colors = plt.cm.tab20c.colors  # Use tab20c colormap

# Create the treemap
plt.figure(figsize=(12,6))
squarify.plot(
    sizes=department_counts_sorted.values,
    label=[f"{dept}\n{count:,}" for dept, count in zip(department_counts_sorted.index, department_counts_sorted.values)],
    color=colors*10,  # Repeat colors if departments > 20
    alpha=0.8
)

plt.title('Employee Distribution by Department', fontsize=18)
plt.axis('off')  # Remove axes
plt.show()

# Average salary per department rounded to 4 decimals
avg_salary_dept = hr_data.groupby('DEPARTMENT')['SALARY_INR'].mean().round(4).sort_values(ascending=False)

# Format output to show as 1.1298 instead of scientific notation
avg_salary_dept_formatted = avg_salary_dept.apply(lambda x: float(f"{x/1e6:.4f}"))  # convert to millions

# Display
print(avg_salary_dept_formatted)

DEPARTMENT
IT            1.1241
Finance       0.9446
R&D           0.7980
Sales         0.7944
Marketing     0.7710
Operations    0.7566
HR            0.7380
Name: SALARY_INR, dtype: float64

import matplotlib.pyplot as plt

# Colors for bars
colors = plt.cm.Pastel2.colors

plt.figure(figsize=(10,6))
bars = plt.bar(avg_salary_dept.index, avg_salary_dept.values / 1e6, color=colors, edgecolor='black')  # in millions

# Show values on top of each bar
for bar in bars:
    height = bar.get_height()
    plt.text(bar.get_x() + bar.get_width()/2, height + 0.01, f'{height:.4f}M', ha='center', va='bottom')

plt.title('Average Salary by Department (in Millions)')
plt.xlabel('Department')
plt.ylabel('Average Salary (M INR)')
plt.xticks(rotation=45)
plt.show()

import squarify

plt.figure(figsize=(12,6))
squarify.plot(
    sizes=avg_salary_dept.values / 1e6,
    label=[f"{dept}\n{salary:.4f}M" for dept, salary in zip(avg_salary_dept.index, avg_salary_dept.values / 1e6)],
    color=colors*10,
    alpha=0.8
)
plt.title('Average Salary by Department (in Millions)')
plt.axis('off')
plt.show()

# Average salary per Job Title
avg_salary_job = hr_data.groupby('JOB_TITLE')['SALARY_INR'].mean()

# Find the job title with the highest average salary
highest_salary_job = avg_salary_job.idxmax()
highest_salary_value = avg_salary_job.max()

print(f"Job Title with Highest Average Salary: {highest_salary_job}")
print(f"Average Salary: {highest_salary_value:,.2f} INR")

Job Title with Highest Average Salary: IT Manager
Average Salary: 2,091,632.61 INR

import matplotlib.pyplot as plt
import matplotlib

# Top 10 highest paying jobs
top_10_jobs = avg_salary_job.sort_values(ascending=False).head(10)

# Normalize salary values for color intensity
norm = matplotlib.colors.Normalize(vmin=min(top_10_jobs.values), vmax=max(top_10_jobs.values))
cmap = plt.cm.plasma  # You can try 'viridis', 'cividis', 'coolwarm'
colors = [cmap(norm(value)) for value in top_10_jobs.values]

# Plot
plt.figure(figsize=(12,6))
bars = plt.barh(top_10_jobs.index[::-1], top_10_jobs.values[::-1]/1e6, color=colors, edgecolor='black')  # in millions

# Show salary values on bars
for bar, value in zip(bars, top_10_jobs.values[::-1]/1e6):
    plt.text(bar.get_width() + 0.01, bar.get_y() + bar.get_height()/2, f'{value:.4f}M', va='center')

plt.xlabel('Average Salary (M INR)')
plt.ylabel('Job Title')
plt.title('Top 10 Highest Paying Job Titles')
plt.show()

# Top 10 highest paying jobs
top_10_jobs_list = avg_salary_job.sort_values(ascending=False).head(10).index

# Filter dataset for these jobs
top_jobs_data = hr_data[hr_data['JOB_TITLE'].isin(top_10_jobs_list)]

# Group by JOB_TITLE and WORK_MODE, then calculate average salary
stacked_data = top_jobs_data.groupby(['JOB_TITLE', 'WORK_MODE'])['SALARY_INR'].mean().unstack(fill_value=0)

# Display
print(stacked_data)

WORK_MODE                          On-site        Remote
JOB_TITLE                                               
Brand Manager                 8.168924e+05  7.904691e+05
Business Development Manager  1.248739e+06  1.252621e+06
Content Strategist            8.166190e+05  7.890550e+05
Finance Manager               1.748182e+06  1.756589e+06
Financial Analyst             1.053618e+06  1.063148e+06
HR Manager                    1.234580e+06  1.286108e+06
IT Manager                    2.094311e+06  2.087500e+06
Innovation Manager            8.073284e+05  8.212004e+05
Sales Director                7.947794e+05  8.192545e+05
Software Engineer             1.190134e+06  1.196537e+06

import matplotlib.pyplot as plt

# Assuming your DataFrame is named stacked_data
# (if it's named differently, replace stacked_data with your df name)

# Colors for work modes
colors = ['skyblue', 'orange']

# Plot stacked bar chart
ax = stacked_data.plot(kind='bar',
                       stacked=True,
                       figsize=(12, 6),
                       color=colors,
                       edgecolor='black')

# Add labels inside bars
for i in ax.patches:
    width = i.get_width()
    height = i.get_height()
    x, y = i.get_xy()

    if height > 0:
        ax.text(x + width/2,
                y + height/2,
                f'{height/1e6:.2f}M',   # convert to millions with 2 decimals
                ha='center',
                va='center',
                fontsize=8,
                color='black')

plt.ylabel('Average Salary (INR in Millions)')
plt.xlabel('Job Title')
plt.title('Average Salary by Job Title and Work Mode (Stacked)')
plt.xticks(rotation=90)
plt.legend(title='Work Mode')
plt.tight_layout()
plt.show()

# Group by Department and Job Title, calculate average salary
dept_job_salary = hr_data.groupby(['DEPARTMENT', 'JOB_TITLE'])['SALARY_INR'].mean().unstack(fill_value=0)

# Display the top rows
dept_job_salary.head()

import matplotlib.pyplot as plt

plt.figure(figsize=(15,7))
colors = plt.cm.tab20.colors

bars = dept_job_salary.plot(
    kind='bar',
    stacked=True,
    figsize=(15,7),
    color=colors,
    edgecolor='black'
)

# Add values on each segment
for i, dept in enumerate(dept_job_salary.index):
    bottom = 0
    for j, job in enumerate(dept_job_salary.columns):
        height = dept_job_salary.loc[dept, job]
        if height > 0:
            plt.text(
                i,
                bottom + height/2,
                f'{height/1e6:.4f}M',  # Show in millions
                ha='center',
                va='center',
                fontsize=8,
                color='black'
            )
        bottom += height

plt.ylabel('Average Salary (INR)')
plt.xlabel('Department')
plt.title('Average Salary in Departments by Job Title (with Values)')
plt.xticks(rotation=45)
plt.legend(title='Job Title', bbox_to_anchor=(1, 1))
plt.show()

<Figure size 1500x700 with 0 Axes>

# Filter dataset
resigned_terminated = hr_data[hr_data['STATUS'].isin(['Resigned', 'Terminated'])]

# Group by Department and Status
dept_status_counts = resigned_terminated.groupby(['DEPARTMENT', 'STATUS']).size().unstack(fill_value=0)

# Display
print(dept_status_counts)

STATUS      Resigned  Terminated
DEPARTMENT                      
Finance         1756         406
HR              1383         333
IT              5067        1308
Marketing       2076         531
Operations      2611         646
R&D              860         222
Sales           3452         879

import matplotlib.pyplot as plt

colors = ['orange', 'red']  # Resigned = orange, Terminated = red

# Plot
dept_status_counts.plot(kind='bar', stacked=True, figsize=(12,6), color=colors, edgecolor='black')

# Add values on bars
for i, dept in enumerate(dept_status_counts.index):
    bottom = 0
    for j, status in enumerate(dept_status_counts.columns):
        height = dept_status_counts.loc[dept, status]
        if height > 0:
            plt.text(
                i,
                bottom + height/2,
                f'{height:,}',  # Show count with comma
                ha='center',
                va='center',
                fontsize=10,
                color='black'
            )
        bottom += height

plt.ylabel('Number of Employees')
plt.xlabel('Department')
plt.title('Number of Resigned & Terminated Employees by Department')
plt.xticks(rotation=45)
plt.legend(title='Status')
plt.show()

sample_data = hr_data.sample(5000, random_state=42)  # 5000 rows sample

plt.figure(figsize=(14,7))
sns.scatterplot(
    data=sample_data,
    x='EXPERIENCE_YEARS',
    y=sample_data['SALARY_INR']/1e6,
    hue='DEPARTMENT',
    palette='tab10',
    alpha=0.6,
    s=20
)
plt.xlabel('Years of Experience')
plt.ylabel('Salary (M INR)')
plt.title('Salary vs Years of Experience by Department (Sampled)')
plt.legend(title='Department', bbox_to_anchor=(1,1))
plt.show()

# Convert PERFORMANCE_RATING to numeric (if not already)
hr_data['PERFORMANCE_RATING'] = pd.to_numeric(hr_data['PERFORMANCE_RATING'], errors='coerce')

# Group by Department and calculate mean
avg_perf_dept = hr_data.groupby('DEPARTMENT')['PERFORMANCE_RATING'].mean().sort_values(ascending=False)

# Round to 2 decimals
avg_perf_dept = avg_perf_dept.round(2)

# Display
print(avg_perf_dept)

DEPARTMENT
Finance       3.03
IT            3.01
Marketing     3.01
Sales         3.01
Operations    2.99
R&D           2.98
HR            2.98
Name: PERFORMANCE_RATING, dtype: float64

plt.figure(figsize=(12,6))
bars = plt.barh(avg_perf_dept.index, avg_perf_dept.values, color='orange', edgecolor='black')

# Add values on bars
for bar, value in zip(bars, avg_perf_dept.values):
    plt.text(value + 0.02, bar.get_y() + bar.get_height()/2, f'{value}', va='center', fontsize=10)

plt.xlabel('Average Performance Rating')
plt.ylabel('Department')
plt.title('Average Performance Rating by Department')
plt.show()

# Count employees per country
country_counts = hr_data['LOCATION'].value_counts()

# Display top countries
print(country_counts.head(10))

LOCATION
East Michael, Kenya                             4
South Matthew, Zambia                           4
South David, Bhutan                             4
Johnsonchester, Antigua and Barbuda             3
South John, Costa Rica                          3
New Jason, Heard Island and McDonald Islands    3
Jamesside, Poland                               3
Johnfort, Korea                                 3
Charlesbury, Turks and Caicos Islands           3
Michaelmouth, Swaziland                         3
Name: count, dtype: int64

import pandas as pd

# Count employees per country
country_counts = hr_data['LOCATION'].value_counts().reset_index()
country_counts.columns = ['Country', 'Employees']

# Display top rows
country_counts.head()

import plotly.express as px

fig = px.choropleth(
    country_counts,
    locations='Country',       # Country names
    locationmode='country names',
    color='Employees',         # Column to determine color
    hover_name='Country',
    color_continuous_scale='Blues',
    title='Employee Distribution by Country'
)

fig.show()

/tmp/ipython-input-3641378741.py:3: DeprecationWarning:

The library used by the *country names* `locationmode` option is changing in an upcoming version. Country names in existing plots may not work in the new version. To ensure consistent behavior, consider setting `locationmode` to *ISO-3*.

# Count employees by Department and Country (only top 5 countries for readability)
top_countries_list = country_counts['Country'].head(5)
dept_country_counts = hr_data[hr_data['LOCATION'].isin(top_countries_list)].groupby(['LOCATION','DEPARTMENT']).size().unstack(fill_value=0)

# Plot stacked area
dept_country_counts.T.plot.area(figsize=(12,6), alpha=0.7)
plt.ylabel('Number of Employees')
plt.xlabel('Department')
plt.title('Department-wise Employee Distribution in Top Countries (Stacked Area)')
plt.show()

# Convert columns to numeric if not already
hr_data['SALARY_INR'] = pd.to_numeric(hr_data['SALARY_INR'], errors='coerce')
hr_data['PERFORMANCE_RATING'] = pd.to_numeric(hr_data['PERFORMANCE_RATING'], errors='coerce')

# Drop rows with missing values
df_clean = hr_data.dropna(subset=['SALARY_INR', 'PERFORMANCE_RATING'])

correlation = df_clean['SALARY_INR'].corr(df_clean['PERFORMANCE_RATING'])
print(f"Correlation between Salary and Performance Rating: {correlation:.4f}")

Correlation between Salary and Performance Rating: 0.0016

import matplotlib.pyplot as plt
import seaborn as sns

plt.figure(figsize=(12,6))

sns.regplot(
    data=df_clean,
    x='PERFORMANCE_RATING',
    y=df_clean['SALARY_INR']/1e6,   # Convert to millions
    scatter_kws={'alpha':0.4, 's':30, 'color':'blue'},  # lighter, smaller points
    line_kws={'color':'red', 'lw':2}  # bold red trend line
)

plt.xlabel('Performance Rating')
plt.ylabel('Salary (Millions INR)')
plt.title('Correlation between Salary and Performance Rating', fontsize=14, fontweight='bold')
plt.grid(True, linestyle='--', alpha=0.6)  # add light grid
plt.show()

# Ensure HIRE_DATE is datetime
hr_data['HIRE_DATE'] = pd.to_datetime(hr_data['HIRE_DATE'], errors='coerce')

# Extract year
hr_data['HIRE_YEAR'] = hr_data['HIRE_DATE'].dt.year

# Count hires per year
hires_per_year = hr_data['HIRE_YEAR'].value_counts().sort_index()
print(hires_per_year.head())

HIRE_YEAR
2010.0     662
2011.0    1770
2012.0    1731
2013.0    1694
2014.0    1803
Name: count, dtype: int64

import matplotlib.pyplot as plt

plt.figure(figsize=(12,6))
plt.plot(hires_per_year.index, hires_per_year.values, marker='o', linestyle='-', color='blue')

plt.title('Number of Hires per Year')
plt.xlabel('Year')
plt.ylabel('Number of Hires')
plt.grid(True, alpha=0.3)
plt.show()

plt.figure(figsize=(12,6))
plt.fill_between(hires_per_year.index, hires_per_year.values, color='skyblue', alpha=0.6)
plt.plot(hires_per_year.index, hires_per_year.values, color='blue', linewidth=2)

plt.title('Hiring Trend Over the Years (Area Chart)')
plt.xlabel('Year')
plt.ylabel('Number of Hires')
plt.grid(True, alpha=0.3)
plt.show()

# Group salaries by work mode
salary_workmode = hr_data.groupby('WORK_MODE')['SALARY_INR'].describe()

print(salary_workmode[['mean', '50%', 'min', 'max']])

                    mean       50%       min        max
WORK_MODE                                              
On-site    895233.270432  810544.0  300260.0  2997010.0
Remote     894996.884833  810172.0  300255.0  2994359.0

import matplotlib.pyplot as plt
import seaborn as sns

plt.figure(figsize=(10,6))
sns.boxplot(
    data=hr_data,
    x='WORK_MODE',
    y=hr_data['SALARY_INR']/1e6,   # in millions
    palette='Set2'
)
plt.title('Salary Distribution: Remote vs On-site')
plt.ylabel('Salary (M INR)')
plt.xlabel('Work Mode')
plt.show()

/tmp/ipython-input-2898040175.py:5: FutureWarning:


Passing `palette` without assigning `hue` is deprecated and will be removed in v0.14.0. Assign the `x` variable to `hue` and set `legend=False` for the same effect.

plt.figure(figsize=(10,6))
sns.violinplot(
    data=hr_data,
    x='WORK_MODE',
    y=hr_data['SALARY_INR']/1e6,
    palette='Pastel1'
)
plt.title('Salary Distribution by Work Mode (Violin Plot)')
plt.ylabel('Salary (M INR)')
plt.xlabel('Work Mode')
plt.show()

/tmp/ipython-input-1407532913.py:2: FutureWarning:


Passing `palette` without assigning `hue` is deprecated and will be removed in v0.14.0. Assign the `x` variable to `hue` and set `legend=False` for the same effect.

from scipy.stats import ttest_ind

remote_salaries = hr_data.loc[hr_data['WORK_MODE']=='Remote', 'SALARY_INR']
onsite_salaries = hr_data.loc[hr_data['WORK_MODE']=='On-site', 'SALARY_INR']

t_stat, p_value = ttest_ind(remote_salaries, onsite_salaries, equal_var=False)

print(f"T-statistic: {t_stat:.4f}, P-value: {p_value:.4e}")

T-statistic: -0.0852, P-value: 9.3206e-01

# Sort by department and salary
hr_data_sorted = hr_data.sort_values(['DEPARTMENT', 'SALARY_INR'], ascending=[True, False])

# Pick top 10 per department
top10_per_dept = hr_data_sorted.groupby('DEPARTMENT').head(10)

# Show selected columns
top10_per_dept[['DEPARTMENT', 'FULL_NAME', 'JOB_TITLE', 'SALARY_INR']].head(30)

# Add rank within each department
hr_data['Salary_Rank'] = hr_data.groupby('DEPARTMENT')['SALARY_INR'].rank(method='first', ascending=False)

# Filter top 10
top10_per_dept = hr_data[hr_data['Salary_Rank'] <= 10].sort_values(['DEPARTMENT', 'Salary_Rank'])

top10_per_dept[['DEPARTMENT', 'Salary_Rank', 'FULL_NAME', 'JOB_TITLE', 'SALARY_INR']].head(30)

import matplotlib.pyplot as plt

it_top10 = top10_per_dept[top10_per_dept['DEPARTMENT']=='IT']

plt.figure(figsize=(12,6))
plt.barh(it_top10['FULL_NAME'], it_top10['SALARY_INR']/1e6, color='skyblue')
plt.xlabel('Salary (M INR)')
plt.title('Top 10 Salaries in IT Department')
plt.gca().invert_yaxis()
plt.show()

# Total employees per department
dept_total = hr_data.groupby('DEPARTMENT')['STATUS'].count()

# Resigned employees per department
dept_resigned = hr_data[hr_data['STATUS']=='Resigned'].groupby('DEPARTMENT')['STATUS'].count()

# Combine
attrition_rate = (dept_resigned / dept_total * 100).sort_values(ascending=False)

# Display
print(attrition_rate.head(10).round(2))

DEPARTMENT
R&D           20.26
Finance       20.11
Sales         20.05
Marketing     19.91
Operations    19.81
HR            19.70
IT            19.54
Name: STATUS, dtype: float64

import matplotlib.pyplot as plt

plt.figure(figsize=(12,6))
bars = plt.bar(attrition_rate.index, attrition_rate.values, color='coral', edgecolor='black')

# Add values on bars
for bar, value in zip(bars, attrition_rate.values):
    plt.text(bar.get_x() + bar.get_width()/2, bar.get_height() + 0.5,
             f'{value:.2f}%', ha='center', fontsize=10)

plt.ylabel('Attrition Rate (%)')
plt.xlabel('Department')
plt.title('Attrition Rate (Resigned %) by Department')
plt.xticks(rotation=45)
plt.show()

plt.figure(figsize=(12,6))
bars = plt.barh(attrition_rate.index, attrition_rate.values, color='tomato', edgecolor='black')

# Add values
for bar, value in zip(bars, attrition_rate.values):
    plt.text(value + 0.5, bar.get_y() + bar.get_height()/2,
             f'{value:.2f}%', va='center', fontsize=10)

plt.xlabel('Attrition Rate (%)')
plt.ylabel('Department')
plt.title('Attrition Rate (Resigned %) by Department')
plt.gca().invert_yaxis()
plt.show()

import pandas as pd

# Create target variable
hr_data['Attrition'] = hr_data['STATUS'].apply(lambda x: 1 if x=='Resigned' else 0)

features = ['DEPARTMENT','JOB_TITLE','EXPERIENCE_YEARS','PERFORMANCE_RATING','SALARY_INR','WORK_MODE','LOCATION']
X = hr_data[features]
y = hr_data['Attrition']

from sklearn.model_selection import train_test_split
from sklearn.preprocessing import LabelEncoder

# Encode categorical variables
X = X.copy()
for col in ['DEPARTMENT','JOB_TITLE','WORK_MODE','LOCATION']:
    X[col] = LabelEncoder().fit_transform(X[col].astype(str))

# Train-test split
X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.2, random_state=42)

from sklearn.ensemble import RandomForestClassifier
from sklearn.metrics import accuracy_score, classification_report, confusion_matrix

# Train
model = RandomForestClassifier(n_estimators=100, random_state=42)
model.fit(X_train, y_train)

# Predict
y_pred = model.predict(X_test)

# Evaluation
print("Accuracy:", accuracy_score(y_test, y_pred))
print("\nClassification Report:\n", classification_report(y_test, y_pred))

Accuracy: 0.7886577142527809

Classification Report:
               precision    recall  f1-score   support

           0       0.80      0.98      0.88     13901
           1       0.21      0.02      0.04      3450

    accuracy                           0.79     17351
   macro avg       0.51      0.50      0.46     17351
weighted avg       0.68      0.79      0.71     17351

import matplotlib.pyplot as plt

importances = model.feature_importances_
feat_importances = pd.Series(importances, index=X.columns)
feat_importances.nlargest(10).plot(kind='barh', color='teal')
plt.title("Top Features Influencing Attrition")
plt.show()

# Step 1: Upload your .ipynb file
from google.colab import files
uploaded = files.upload()   # Choose HR_Data_MNC.ipynb from your computer

# Step 2: Convert uploaded notebook to HTML
import os
for fn in uploaded.keys():
    os.system(f"jupyter nbconvert --to html '{fn}'")

# Step 3: Download the converted HTML file
for fn in uploaded.keys():
    html_file = fn.replace(".ipynb", ".html")
    files.download(html_file)

	Unnamed: 0	Employee_ID	Full_Name	Department	Job_Title	Hire_Date	Location	Performance_Rating	Experience_Years	Status	Work_Mode	Salary_INR
0	0	EMP0000001	Joshua Nguyen	IT	Software Engineer	2011-08-10	Isaacland, Denmark	5.0	14.0	Resigned	On-site	1585363.0
1	1	EMP0000002	Julie Williams	Marketing	SEO Specialist	2018-03-02	Anthonyside, Costa Rica	2.0	7.0	Active	On-site	847686.0
2	2	EMP0000003	Alyssa Martinez	HR	HR Manager	2023-03-20	Port Christinaport, Saudi Arabia	1.0	2.0	Active	On-site	1430084.0
3	3	EMP0000004	Nicholas Valdez	IT	Software Engineer	2023-10-12	Port Shelbychester, Antigua and Barbuda	1.0	1.0	Active	On-site	990689.0
4	4	EMP0000005	Joel Hendricks	Operations	Logistics Coordinator	2024-12-09	Lake Kimberly, Palestinian Territory	5.0	0.0	Active	On-site	535082.0
5	5	EMP0000006	Jason Gardner	Operations	Logistics Coordinator	2021-02-23	Zimmermanstad, Bulgaria	5.0	4.0	Active	On-site	641393.0
6	6	EMP0000007	Julie Wright	Finance	Finance Manager	2016-04-04	Karenfort, South Africa	2.0	9.0	Active	On-site	1383891.0
7	7	EMP0000008	Scott Wilson	Sales	Account Manager	2020-04-04	Moniqueview, British Virgin Islands	2.0	5.0	Active	Remote	423091.0
8	8	EMP0000009	Cathy Thompson	Finance	Financial Analyst	2018-05-29	South Catherine, Belize	4.0	7.0	Resigned	Remote	1138452.0
9	9	EMP0000010	Maria Yu MD	IT	Software Engineer	2015-10-08	Brownport, Yemen	4.0	9.0	Active	Remote	1543102.0

	UNNAMED: 0	EMPLOYEE_ID	FULL_NAME	DEPARTMENT	JOB_TITLE	HIRE_DATE	LOCATION	PERFORMANCE_RATING	EXPERIENCE_YEARS	STATUS	WORK_MODE	SALARY_INR
0	0	EMP0000001	Joshua Nguyen	IT	Software Engineer	2011-08-10	Isaacland, Denmark	5.0	14.0	Resigned	On-site	1585363.0
1	1	EMP0000002	Julie Williams	Marketing	SEO Specialist	2018-03-02	Anthonyside, Costa Rica	2.0	7.0	Active	On-site	847686.0
2	2	EMP0000003	Alyssa Martinez	HR	HR Manager	2023-03-20	Port Christinaport, Saudi Arabia	1.0	2.0	Active	On-site	1430084.0
3	3	EMP0000004	Nicholas Valdez	IT	Software Engineer	2023-10-12	Port Shelbychester, Antigua and Barbuda	1.0	1.0	Active	On-site	990689.0
4	4	EMP0000005	Joel Hendricks	Operations	Logistics Coordinator	2024-12-09	Lake Kimberly, Palestinian Territory	5.0	0.0	Active	On-site	535082.0
...	...	...	...	...	...	...	...	...	...	...	...	...
86749	86749	EMP0086750	Heather Price	HR	HR Director	2018-03-24	Lake Josephberg, Sierra Leone	4.0	7.0	Active	Remote	1019554.0
86750	86750	EMP0086751	Robert Jones	Finance	Financial Analyst	2024-06-18	Erinmouth, Tuvalu	2.0	1.0	Active	Remote	1465772.0
86751	86751	EMP0086752	Erica Brown	Sales	Sales Executive	2015-11-13	South Virginia, Christmas Island	5.0	9.0	Active	On-site	874900.0
86752	86752	EMP0086753	Kathleen Mccormick	IT	Software Engineer	2025-02-06	Lake Terri, Cayman Islands	5.0	0.0	Retired	Remote	1388124.0
86753	86753	EMP0086754	Rebecca Wilson	R&D	Lab Technician	NaN	NaN	NaN	NaN	NaN	NaN	NaN

	SI.	PERFORMANCE_RATING	EXPERIENCE_YEARS	SALARY_INR
count	86754.000000	86753.000000	86753.000000	8.675300e+04
mean	43376.500000	3.004784	5.026097	8.951380e+05
std	25043.866964	1.414780	3.615067	4.007010e+05
min	0.000000	1.000000	0.000000	3.002550e+05
25%	21688.250000	2.000000	2.000000	6.154810e+05
50%	43376.500000	3.000000	5.000000	8.103920e+05
75%	65064.750000	4.000000	8.000000	1.072905e+06
max	86753.000000	5.000000	15.000000	2.997010e+06

	count
STATUS
Active	60770
Resigned	17205
Retired	4453
Terminated	4325

JOB_TITLE	Account Manager	Accountant	Brand Manager	Business Development Manager	CFO	CTO	Content Strategist	Data Analyst	DevOps Engineer	Finance Manager	...	Operations Director	Operations Executive	Product Developer	Research Scientist	SEO Specialist	Sales Director	Sales Executive	Software Engineer	Supply Chain Manager	Talent Acquisition Specialist
DEPARTMENT
Finance	0.0	651868.153082	0.000000	0.0	791215.349892	0.000000	0.000000	0.000000	0.000000	1.751508e+06	...	0.000000	0.000000	0.0	0.0	0.000000	0.0	0.0	0.000000e+00	0.000000	0.000000
HR	0.0	0.000000	0.000000	0.0	0.000000	0.000000	0.000000	0.000000	0.000000	0.000000e+00	...	0.000000	0.000000	0.0	0.0	0.000000	0.0	0.0	0.000000e+00	0.000000	786442.640772
IT	0.0	0.000000	0.000000	0.0	0.000000	798643.820974	0.000000	795984.510555	798789.399897	0.000000e+00	...	0.000000	0.000000	0.0	0.0	0.000000	0.0	0.0	1.192761e+06	0.000000	0.000000
Marketing	0.0	0.000000	806351.676895	0.0	0.000000	0.000000	805044.184741	0.000000	0.000000	0.000000e+00	...	0.000000	0.000000	0.0	0.0	701029.080287	0.0	0.0	0.000000e+00	0.000000	0.000000
Operations	0.0	0.000000	0.000000	0.0	0.000000	0.000000	0.000000	0.000000	0.000000	0.000000e+00	...	799731.567742	803170.514597	0.0	0.0	0.000000	0.0	0.0	0.000000e+00	798376.231964	0.000000

HR Data Analytics for Business Decision-Making (MNC Project)

Importing Libraries¶

1. What is the distribution of Employee Status (Active, Resigned, Retired, Terminated)?¶

2. What is the distribution of work modes (On-site, Remote)?¶

3. How many employees are there in each department?¶

4. What is the average salary by Department?¶

5. Which job title has the highest average salary?¶

6. What is the average salary in different Departments based on Job Title?¶

7. How many employees resigned & terminated in each department?¶

8. How does salary vary with years of experience?¶

9. What is the average performance rating by department?¶

10. Which Country has the highest concentration of employees?¶

11. Is there a correlation between performance rating and salary?¶

12. How has the number of hires changed over time (per year) ?¶

13. Compare salaries of Remote vs. On-site employees — is there a significant difference ?¶

If p < 0.05 → difference is statistically significant.¶

If p ≥ 0.05 → no strong evidence of difference.¶

14. Find the top 10 employees with the highest salary in each department.¶

15. Identify departments with the highest attrition rate (Resigned %).¶

16. Train a Model (Random Forest Classifier)¶

Step 1: Choose Target Variable¶

Step 2: Feature Selection¶

Step 3: Preprocessing (Encode Categorical Data)¶

Step 4: Train a Model (Random Forest Classifier)¶

Step 5: Feature Importance¶

📌 Final Conclusion¶

	0
Unnamed: 0	86754
Employee_ID	86754
Full_Name	86754
Department	86754
Job_Title	86754
Hire_Date	86753
Location	86753
Performance_Rating	86753
Experience_Years	86753
Status	86753
Work_Mode	86753
Salary_INR	86753

	Country	Employees
0	East Michael, Kenya	4
1	South Matthew, Zambia	4
2	South David, Bhutan	4
3	Johnsonchester, Antigua and Barbuda	3
4	South John, Costa Rica	3

	DEPARTMENT	FULL_NAME	JOB_TITLE	SALARY_INR
73158	Finance	Veronica Perez	Finance Manager	2498261.0
54260	Finance	Michelle Bryant	Finance Manager	2494985.0
61658	Finance	Cassandra Blake	Finance Manager	2493293.0
71681	Finance	Crystal Edwards	Finance Manager	2492436.0
8989	Finance	Nicholas Cuevas	Finance Manager	2490595.0
14262	Finance	Susan Romero	Finance Manager	2489616.0
32099	Finance	Connie Coleman	Finance Manager	2488951.0
27788	Finance	Krista Perez	Finance Manager	2487648.0
22810	Finance	Samantha Mcmahon	Finance Manager	2487602.0
25563	Finance	Jimmy Foster	Finance Manager	2487200.0
61738	HR	Cole Jackson	HR Manager	1798728.0
21072	HR	Kenneth Morris	HR Manager	1798687.0
21408	HR	Sarah Arnold	HR Manager	1798020.0
84818	HR	Brendan Bush	HR Manager	1797326.0
84041	HR	Melinda Gutierrez	HR Manager	1796885.0
13064	HR	Shane Hale	HR Manager	1795800.0
26001	HR	Deborah Robinson	HR Manager	1794550.0
72406	HR	Jessica Powell	HR Manager	1794426.0
60646	HR	William Richardson	HR Manager	1793603.0
27848	HR	Eric Ramos	HR Manager	1791447.0
45059	IT	Sean Stone	IT Manager	2997010.0
1823	IT	Christina West	IT Manager	2994359.0
39836	IT	Sheena Thomas	IT Manager	2992188.0
48704	IT	Nicole Banks	IT Manager	2991695.0
73065	IT	Sarah Page	IT Manager	2990007.0
54888	IT	John Fitzgerald	IT Manager	2989764.0
21877	IT	Cindy Bryan	IT Manager	2988770.0
50316	IT	Miss Debra Love	IT Manager	2988356.0
7516	IT	Traci Holt	IT Manager	2986637.0
69551	IT	Robin Wang	IT Manager	2985864.0